How to Prep Aluminum for Welding: A Practical Guide

The bead looked fine until you hit the chip hammer and it crumbled into a line of little craters. That silvery dust of aluminum oxide is the whole story: it melts at roughly 3,600°F, while the metal underneath melts at about 1,220°F, so a dirty aluminum plate welds like two separate pieces of metal sitting next to each other. Prepping aluminum is not fussy extra steps you can skip when you are in a hurry — it is the single cheapest fix for porosity, lack of fusion, and cracked joints you will ever do.

Why Aluminum Fails Before the Arc Even Strikes

The bead looked picture perfect going down the joint. A pass or two later, when you go to knock at it with a chip hammer, it crumbles into a line of little craters. More often than not, that failure has nothing to do with your amperage, gas flow, or technique. It is the invisible layer of oxidation sitting on the metal before your arc ever strikes.

Before any power is turned on, there are only two prep operations that can prevent the majority of these failures: degreasing and oxide removal. Both exist for a physical reason, and skipping either one is gambling with the joint. For a deeper breakdown of how these tie into machine setup, it is worth brushing up on aluminum welding basics before moving any further.

The Oxide Barrier

The biggest offender is the naturally forming oxide layer. When exposed to air, aluminum instantly reacts to form aluminum oxide (alumina, Al₂O₃). This film is extremely thin on clean, dry material, measured in angstroms, and it is self-limiting. The problem is its melting point. Aluminum oxide melts at approximately 3,600°F, while the base metal underneath only melts at about 1,220°F (a value also listed by OSHA’s Occupational Chemical Database for aluminum metal). That is a difference of over 2,300 degrees.

What this means practically is brutal: by the time your puddle is hot enough to melt the aluminum plate, that oxide is still solid. Instead of mixing together, the molten base metal tries to wet around chunks of unmelted alumina. The result is a weak, porous, and extremely brittle weld. On top of this, the oxide reforms almost instantly in open air. Even after brushing it off, it will begin reforming within minutes, so timing between cleaning and welding is just as important as the cleaning itself.

Why Aluminum Punishes Lazy Prep

The oxide is only half of the fight. Aluminum is uniquely unforgiving for a handful of other reasons that amplify any bit of contamination left behind:

• Insane heat transfer: Aluminum conducts heat roughly four to five times faster than mild steel. This pulls heat away from the joint at a blistering rate, making it much harder to get fusion without dumping excess amperage.
• Low melting point: With a melting point around 1,220°F depending on the alloy, it is incredibly easy to overheat, burn through thin stock, or warp it with poor heat control.
• High shrinkage: As it cools, aluminum shrinks significantly. Without proper fit-up and fixturing, that shrinkage will pull joints out of alignment every time.
• Hydrogen and moisture sensitivity: Aluminum has a strong tendency to trap hydrogen and moisture. Oils, fingerprints, shop humidity, or even a damp work surface can introduce that moisture directly into the puddle, leading to porosity.

Not all oxidation is equal either. A barely visible film from fresh stock is one thing, but hydrated/white oxide (a chalky, frosty white appearance) is far more porous and moisture-absorbent. This type of oxidation requires true mechanical cleaning, not just a quick wipe down.

The Real Cost of Skipping Prep

From an audit standpoint, failed aluminum welds cost far more than the time it takes to clean them. Re-work, radiography, distortion corrections, and scrapped aluminum will almost always outweigh the cost of a clean towel, a dedicated brush, and a proper solvent.

Every shop that gets consistent, porosity-free aluminum welds does the same thing: they respect the two-step process. Degrease to lift oils and contamination, mechanically clean to break through the stubborn alumina, and they do both with the knowledge of exactly why they’re doing them. For more specifics on the products used for this exact task, take a look at aluminum cleaner for welding.

Put simply, you aren’t “cleaning for looks”. You are clearing a 3,600°F refractory barrier off of metal that wants to melt at 1,220°F. Get that barrier out of the way, and the arc finally has something to fuse. Leave it, and you’re welding two pieces of crusted oxide to each other, and it will crumble every single time.

The Two-Step Prep Method Every Aluminum Weld Should Follow

Most fixes for a bad aluminum weld trace back to the minute or two before the arc struck. Yet shops often treat prep as something you do “if there’s time.” The framework this guide expands: degrease, remove the oxide, prepare the joint geometry, then tack. Four steps, and the first two do the heavy lifting.

Step 1: Degrease

Wipe the joint area and surrounding metal with a solvent that leaves nothing behind. That clears machine oil, cutting fluid, fingerprints, lube from the brake press, the lanolin shop rags pick up off equipment, and moisture. Oils and greases are the most common cause of aluminum porosity: they create hydrogen gas as the metal melts, and hydrogen gas is looking for any excuse to escape.

Step 2: Remove the Oxide

Break the oxide layer with a dedicated stainless steel wire brush, working at roughly 90 degrees to the surface with light pressure, along the grain. That takes a chalky, stained surface back to bright metal. ESAB draws the distinction clearly: a hydrated oxide layer, the white or stained kind, must be mechanically removed, while the thin oxide film on clean, dry material is usually broken up sufficiently by the arc itself.

Step 3: Joint Prep, Step 4: Tack

With the metal clean, move to geometry: bevels, root gaps, fit-up, and clamping. Then tack with the same filler and settings you will run for the root pass. Details come later in this guide.

Why the Order Is Non-Negotiable

Degreasing first is not a preference, it is chemistry. Brush a greasy surface and you do two bad things at once: the abrasive drag smears the oil into a thin, even film instead of removing it, and the brushing heats the surface just enough to help that film wet the fresh metal again. You end up with clean metal under a re-wiped layer of grease, and you have already spent the time. Degrease first and the brush rides on dry metal, so it removes oxide instead of rolling grease into it.

Degreasing after brushing has its own trap. Freshly brushed aluminum is raw, high-energy metal, and a solvent wipe on it can leave a moisture film or solvent residue that re-wets the metal you just cleaned. The sequence protects the surface you created in step 2.

When Degreasing Alone Is Enough

You do not have to mechanically clean every piece of aluminum that comes through the door. On clean, dry stock that is freshly machined, extruded, or cut, the oxide is thin and unhydrated, and the arc itself breaks it. AC TIG does it with the electrode-positive half of the cycle, where the tungsten is positive and the work negative, so cathodic action at the work surface cleans the oxide. Reverse-polarity MIG (DCEN, work negative) gives the same effect.

Stained, weathered, or white chalky oxide is a different material: hydrated oxide is thicker, softer, and tenacious. It will not reliably break under the arc, so it always has to come off mechanically first. Unsure which one you have? Look at it. Gritty, dull, or chalky white means brush it.

Material condition	What the surface looks like	Prep level required
New sheet or extrusion	Bright, even, dry; thin oxide only	Degrease only in most cases. Brush if spec or code requires it.
Stored sheet (racked, wrapped)	Dull with shop dust and handling marks	Degrease and brush the weld face. Dust alone can cause porosity.
Weathered or stained	Chalky white, blotchy, gritty	Degrease and mechanically remove the hydrated oxide. Always.
Post machine shop	Possible cutting oil, coolant, or lube film	Degrease thoroughly, then brush. Watch for coolant trapped under the stock.

Clean Only What You Will Weld

ESAB advises cleaning only slightly larger than the weld face unless otherwise specified. That is about more than efficiency. Every extra square inch brushed is another square inch exposed to shop air, dust, and handling, and oxide starts coming back almost immediately. A tight prep band also gives you a clear visual boundary for inspection and tells the next welder the area was prepared rather than skipped. Cleaning the whole plate feels thorough, but it is slower and opens a bigger window for recontamination. Clean the weld face plus a modest margin, and stop.

Why a Written Prep Routine Pays Off

Aluminum defects are notoriously hard to diagnose after the fact. A bead can look excellent and still be full of porosity underneath, and by then you have burned an hour of argon and your afternoon. A documented prep routine turns “something is wrong” into a repeatable test.

Write down the solvent, the brush, the method, and the time lag between prep and welding. When a joint fails, change only the prep variable and watch what happens. That habit turns guesswork into diagnosis and costs far less than rework, radiography, or a scrapped plate. Our guide to aluminum cleaner for welding goes deeper on the solvent and mechanical side, and the practical guide to aluminum MIG welding covers how the arc behaves once the metal is genuinely clean.

That is the framework: two prep operations, one non-negotiable, then geometry and tacking. The next sections break down the solvents that work, the tools that smear metal instead of cleaning it, and the joint prep that decides whether all that cleaning matters.

Step 1 — Degreasing Aluminum Without Making It Worse

Degreasing is the least glamorous part of aluminum prep, and the most often done badly. The bead looks right, the wire feeds fine, then the joint crumbles under a chipping hammer. The culprit is rarely the oxide but the film of oil, grease, or shop-rag lint on the plate when the arc struck.

Steel’s thin oxide is a nuisance you grind through. Oil on aluminum is a seal: it holds the oxide in place, blocks the arc’s cleaning action, and feeds hydrogen into the pool. That is where porosity comes from, and porosity turns a strong joint into a crumbly one.

ESAB’s guidance on aluminum prep is short: acetone, non-chlorinated brake cleaner, and mild dish soap such as Dawn in hot water, rinsed thoroughly with hot water, are effective aluminum degreasers. Everything else deserves scrutiny.

What to Use

• Acetone. Cuts oil, grease, and fingerprint oils fast, evaporates with no residue, and will not attack aluminum. Highly flammable.
• Non-chlorinated brake cleaner. Cheap, aggressive on grease, dries fast. Confirm the label says non-chlorinated.
• Mild dish soap in hot water. Surprising but effective. Detergent breaks the oil loose, hot water carries it away. The rinse is critical: a thorough hot-water rinse, because detergent residue causes porosity and discoloration.
• Dedicated aluminum degreasers. Non-chlorinated, often low-residue. Worth using on repeat production work where consistency matters more than cost per wipe.

What to Avoid

Chlorinated solvents are the classic no. Residual chloride sets up localized corrosion, and aluminum’s oxide film is attacked by chlorides, leaving a pitted, weakened surface that welds poorly and can corrode for years.

Silicone products are worse. Silicone is nearly impossible to remove once it wicks into the microscopic texture of aluminum, and it travels with your filler rod into the weld. General metal sprays, penetrating oils, aerosol lubricants, kerosene, and heavy oils fit here too: slow to evaporate, they sit in the joint and surface later as pores.

The Rag Problem

More aluminum welds are ruined by the rag than by the solvent. A rag used to wipe machines carries lanolin from waylube and greases; a laundered rag carries detergent residue; a rag near a parts washer or silicone spray carries silicone. None of it shows, and all of it transfers the moment you wipe.

ESAB is direct: apply degreaser with white industrial paper towels and avoid shop rags, which can carry lanolin or detergent residue onto the aluminum. Stock single-use white towels, or towels reserved for aluminum, in their own dispenser. Then wipe the cleaned metal with a dry, unused white towel. If it picks up anything, you are not finished.

Technique

Wipe, do not soak. Soaking leaves a bath of contaminants in the joint that can flash back when the arc strikes. Work in one direction, moving clean solvent across the surface instead of dragging contamination back, and change towels often.

Dry with warm air or clean, dry compressed air, then let the part sit a minute so trapped liquid evaporates from crevices. Sequence the job so the last piece degreased is the first piece welded.

Timing: Degrease Before You Tack

ESAB’s instruction is unambiguous: always degrease before tacking. A tack welds two surfaces together, so if either carries oil, that oil is sealed inside the joint where no later wiping reaches. Every pass on top welds over trapped contamination.

Machining follows the same logic. If edge beveling uses a lubricant, ESAB notes that the part must be degreased again before welding. A lubricated carbide bit or coolant-carrying grinder works the same way: grind, wipe, then weld.

Storage Between Jobs

Freshly degreased aluminum is only clean until somebody touches it. Sheet and plate pick up dust, moisture, and fingerprints sitting on a bench. Keep stock racked and off the floor, and for high-quality work store it wrapped.

Working through a backlog of aluminum parts? Build a dedicated aluminum prep area with its own solvents, towels, brushes, and gloves. Prepped aluminum that leaves that area stays clean until it goes into the joint.

Safety

Acetone, brake cleaner, and most dedicated degreasers are flammable liquids with harmful vapors. Use them with the ventilation the safety data sheet calls for, away from hot work, sparks, and flames. Never weld a part still wet with solvent, keep containers closed, and put oily paper towels in a closed, non-leaking container.

Once the surface is clean, dry, and free of shop residue, the next job is the oxide layer, which takes different tools. That is covered in the MIG-specific prep guidance and in the mechanical cleaning step that follows.

Step 2 — Removing the Aluminum Oxide Layer Mechanically and Chemically

Step 1 of prepping aluminum for welding removes oil. Step 2 removes the oxide, and it decides whether the weld is sound.

You are removing a thin, hard, tenacious layer of aluminum oxide (alumina, Al₂O₃) that reforms on exposed aluminum within minutes. Alumina is harder than the aluminum beneath it, chemically inert, and it stays exactly where it is. NIST thermal-property data on aluminum oxide explains why: it remains structurally stable far beyond the melting point of the base metal, so an arc cannot burn through it the way it burns off loose scale on steel. NIST lists the oxide melting near 3,600°F while aluminum melts around 1,220°F.

Run an arc over aluminum and you are heating a small furnace under a ceramic lid. The metal underneath melts, the oxide floats on top, and the two refuse to combine. The bead looks right but has no metallurgical bond beneath it. That is lack of fusion.

Why Mechanical Cleaning Wins for Most Shops

Two ways to strip the oxide: mechanically with brushes, files, and planers, or chemically with etchants and conversion coatings. For most structural, repair, and fabrication work, mechanical wins because it is fast, forgiving, and leaves no residue. Chemistry adds variables instead: bath temperature, dwell time, a rinse and a drying step, each another chance to introduce moisture or an unintended film. It earns its keep on unreachable or unusually stubborn surfaces.

The Tool Ladder, Lightest to Most Aggressive

Climb only as far as you need; each step up costs time and removes base metal.

1. Dedicated stainless steel wire brush. The everyday first choice for plate, sheet, and extrusion. Use stainless wire, not carbon steel, and watch crimp and temper. ESAB recommends stainless wire wheels with light pressure: lean hard and you flatten the wire and burnish the oxide into a tougher film instead of removing it.
2. Non-abrasive Scotch-Brite-type pad used with a brush. The pad is a carrier, not the abrader: load it with the stainless brush so the wire does the work while the pad holds the surface contour. Useful on rolled or extruded profiles a stiff brush cannot reach.
3. Aggressive hand files, 8–10 TPI per ESAB. For joint areas, root surfaces, and land faces the brush cannot reach. A file shears the oxide off and leaves a bright, toothy surface the arc can key into.
4. Hand planers and plane tools for flat stock. A plane takes a thin, even layer off the surface faster than any brush and leaves a uniform result, which is what large plate needs.
5. Carbide burrs and dedicated aluminum grinding tools for bevels. A carbide burr cuts cleanly through the oxide without loading up the way a wheel does.

What to Avoid, and Why

• Grinding discs and flap discs. They smear aluminum: the metal gums up and polishes the disc, leaving exactly the contaminated surface you set out to eliminate.
• Abrasive pads and sandpaper alone. ESAB advises against them because their binders and resins can leave a residue that contributes to porosity.
• Anything that puts iron on the surface. A carbon steel brush, a file that touched steel, or a bench vise with steel jaws introduces iron that causes porosity and cracking.

The Shavings Versus Dust Rule

ESAB recommends avoiding tools that create dust rather than shavings: shavings can be swept up and recycled, while dust spreads contamination through the air and settles everywhere. Fine aluminum dust is also an inhalation hazard. Use it as a filter on any tool in the drawer. A curled shaving is fine; a fine cloud you are going to breathe is not.

When Chemical Methods Make Sense

Chemical etching is the fallback for heavily oxidized surfaces, castings with a rough skin, and spots a brush cannot reach. AWS notes that aluminum’s oxide layer melts at a higher temperature than the base metal and recommends a stainless steel wire brush or a chemical etchant to clean surfaces before welding. Etching dissolves the oxide uniformly, including where a brush skips. The trade-off is time, chemical handling, and a rinse and dry step that must be right: leftover etchant causes porosity.

Alodine and chemical-film conversion coatings are the aerospace and controlled-environment answer. They create a thin, uniform, well-characterized surface that behaves predictably under the arc. If you are working to aerospace welding standards, a conversion coating is often specified rather than improvised with a brush.

The Timing Window Nobody Warns You About

The oxide begins reforming within minutes of exposure to air, so the surface you cleaned ten minutes ago is not what you think it is.

• Clean immediately before welding. Make mechanical cleaning the last step before the arc strikes. Clean, walk away for ten minutes, then weld, and the oxide has had time to grow back.
• Clean the filler and the contact area too. The rod or wire must be free of oxide, oil, and moisture, and the wire contact area on the gun must be clean. A perfect prep still makes a dirty weld if the filler drags dirt into the puddle.

Joint Prep — Bevels, Root Gaps, Fit-Up and Tacking

Scrub a joint until it shines and still weld a failure. Once the metal is clean, geometry takes over: how the edges were prepared, how wide the gap is, how the parts are held. Aluminum conducts heat four to five times better than mild steel and shrinks more as it solidifies, so heat races away from the joint and the metal pulls out of position as it cools. Hobart Brothers’ aluminum guidance is blunt: when porosity appears, all moisture and hydrocarbons must be eliminated before welding, and the fit-up and joint condition have to be right too.

Edge Preparation: Plate and Extrusion

Thin material gets a square butt; there is no point bevelling 16-gauge sheet when the gap already lets the arc strike through. Thicker plate gets a single-V groove, heavy plate a double-V cut from both sides. The groove lets the arc reach the bottom of the joint and control heat input in ways a square butt cannot, and double-V is faster and lower in distortion on very thick plate, though it takes more beveling work.

Aluminum wants a cleaner, more consistent land than steel, and the land is the flat face left at the bottom of the groove. A wavy land creates uneven fusion and makes the arc hunt between tight and open sections, which is how a bead ends up only kissing the root.

For beveling, ESAB recommends a high-speed carbide bit used without lubricant, because lubricant carries moisture and hydrocarbons into the fresh cut. If any was used, degrease the bevel again.

Extrusions need their own note: new architectural extrusion arrives with a mill or anodized film, and die lines can run the full length of the joint. Grind or wire-brush the joint face back to bright metal, not just the edges, and strip anodized or powder-coated material completely from the weld area. For structural work, ask the supplier about the coating’s effect on the finished joint.

Bevel Angles and Root Gaps

For a V-groove in aluminum, 30 to 45 degrees per side is the common range. Thirty degrees means a narrower groove, less filler, less distortion and less heat, but a longer, more constricted arc path that is harder to fill. The 45-degree groove is easier to fill; the steeper groove is gentler on the plate but harder on the operator. Pick one, write it into your procedure and stay with it.

A common starting point for GTAW butt welds in 1/4 inch material is roughly a 1/8 inch root gap, about 3 mm. Thin sections get more gap, because the arc will not fit through a narrow opening without touching both sides; 1/16 or 1/32 inch sheet wants a gap you can clearly see the arc through, often 1/16 inch or more, with a backing bar holding the root in place.

Too tight is the more common failure: a zero gap leaves the arc nowhere to go, giving you spatter, a stubby wide bead and a root that will not fuse. Too wide and you lose heat that should go into the joint, so you get a cold, sluggish root that looks stuck but did not. Set the gap with spacers and clamp it. Do not eyeball it.

Here is a practical starting table for GTAW butt joints on 5052 or 6061 plate. Treat it as a starting point only and verify it against your own qualified procedure, because different alloys, wire diameters and machines shift the numbers.

Material Thickness	Groove	Bevel Angle	Typical Root Gap
Up to 1/16 in (1.6 mm)	Square butt	None	1/16 in (1.6 mm) with backing
1/16 to 1/8 in (1.6–3 mm)	Square butt	None	1/16 in (1.6 mm)
1/8 to 1/4 in (3–6 mm)	Single-V	30–45°	1/16 in (1.6 mm)
1/4 in (6 mm)	Single-V	30–45°	1/8 in (3 mm)
Over 1/4 in (6 mm)	Single-V or double-V	30–45°	1/8 in (3 mm), wider per procedure

Fit-Up: Why Clamping Is Not Optional

Uniform gaps mean the arc sees the same conditions from one end of the joint to the other. If the gap is 1/16 inch at the start and 3/32 inch at the finish, you are welding two different joints in a row and your amperage, travel speed and filler placement become guesswork.

Use strong clamps and strong fixturing. Aluminum’s high thermal conductivity and significant shrinkage keep the joint under stress the entire time it cools. Thin material can move enough to close a root gap while the weld is still hot, which is how you get a bead with no root reinforcement in the middle, and heavy bar can shift enough to change the fit mid-weld. Set the joint, check the gap at several points along its length and clamp it so it cannot move.

Anything that can act as a heat sink matters here. A heavy steel plate under a thin aluminum sheet pulls heat out of the joint so fast the aluminum barely melts. Copper chill bars solve that for a reason: copper takes heat away far better than steel, which sounds backwards until you realize the goal is to control the heat rate in the aluminum, not to remove heat.

Tacking

Tack with the same prep and the same filler metal as the final weld. If the production weld uses 5356 wire, the tack uses 5356 wire. If the production weld is GTAW, the tack is GTAW. A different filler or process changes the chemistry and heat history of the joint, and you have built a defect into the assembly that you will pay for later.

• Small and well fused. A tack should melt into the base metal and show a visible fusion zone. A cold glob sitting on the surface is a defect waiting to happen.
• Same side as the final weld. Tack on the side you intend to weld from. Tacking on the far side means grinding the tack out afterward, and every grind means re-cleaning.
• Use the settings you will use for the root. For MIG that means the spool gun or push-pull feed if you are running one. AWS Welding Digest notes GMAW users need a spool gun or push-pull feed to feed soft aluminum wire without snarls, and tacking with a setup you will not use in production is no preparation at all. See our Aluminum MIG welding guide for how that feed setup changes the job.
• Leave room to work. Space tacks so you can reach the root with the filler rod without fighting the torch.

When you need to remove a bad tack, grind it out completely back to clean metal, then degrease and re-brush the area before welding over it. A ground-out tack leaves a fresh, rough, oxide-covered surface, exactly the condition you worked so hard to avoid.

Backing Bars

A sacrificial backing bar supports the root, controls the gap and lets you run a hotter, more focused arc without blowing through. Steel and copper are the usual choices and both work. Clean and degrease the bar every time, and keep a dedicated aluminum bar separate from your steel stock so the copper and steel surfaces do not cross-contaminate the work.

Never use galvanized backing against an aluminum root. Zinc vaporizes at welding temperatures and gets drawn into the weld pool, where it causes porosity and cracking. The failure looks random and will send you chasing your gas flow and cleaning routine while the real culprit sits right under the joint. Copper is worth the cost on thin material because it controls heat better than steel, letting you keep a tighter bead without melting through, while steel bar is cheaper, tougher and easier to machine and works fine on thicker sections.

Get the geometry right and everything after this section starts to pay off. Clean metal in a well-fitted joint with a properly sized root gap and a sound tack is the foundation. Preheat is generally unnecessary for most aluminum alloys, so let fit-up and sequencing handle the heat.

Filler Metal, Consumables and Keeping Aluminum Separate

A spotless plate still makes a bad weld if the rod feeding the puddle is dirty. Prep is a system of four parts: the plate, the filler, the torch consumables, and the work environment. Clean three and contaminate the fourth and the job only gets harder. Confusing failures follow, like a joint prepped perfectly that welds full of gas pockets because the rod sat in a dusty box or was handled with greasy gloves.

Aluminum rod and wire must be stored dry, clean, and free of oil. Use a heated rod oven, a closed pipe, or a sealed container that keeps dust and moisture out. Rod left open in a shop for a week is not filler metal, it is a moisture collector.

Handling matters as much as storage. Bare hands carry oil, and a glove that has touched a machine carries more. Handle rod with clean, dry gloves, never on a bench or plate. If you contaminate it, a wipe is not enough: heat it or cut off the affected length.

The failure signature is distinctive. Contaminated filler shows up as porosity in the middle of the bead, not at the edges or root. Gas at the toes points to the plate and the shielding gas. Gas dead center points to the rod.

Alloy Selection Is a Separate Decision

Pick the filler for the joint, not for the weld you wish you were making.

• 4043 is the general-purpose rod and wire. It flows well and wets well, covering a wide range of aluminum work, so many shops default to it.
• 5356 is the higher-strength filler and the standard for most heat-treatable alloys and structural work.
• 4043 is generally avoided on alloys with more than about 2 percent silicon, and on joints that get post-weld heat treatment, because of silicon-related hot cracking risk.

None of that changes the prep. A prepped 5356 weld on 6061 can still crack, and a poorly prepped 4043 weld can look fine on a mild job. Filler pairing often gets blamed for contamination when the real culprit was a dirty joint. See our guide to alloy welding procedures and filler metals.

The Dedicated-Tool Discipline

Aluminum tools and steel tools do not mix:

• Separate wire brushes: aluminum or stainless for aluminum, never the reverse.
• Separate hand files and any file used in joint prep.
• Separate grinding wheels, flap discs, and burrs used on bevels.
• Dedicated gloves for aluminum handling, away from general shop gloves.
• A dedicated brush container, so an aluminum brush never sits beside a steel one.

Iron carryover from a steel brush, shared file, or dirty bench will contaminate the weld and can damage a tungsten. For the aluminum-specific cleaning chemistry, see our piece on how to prep aluminum right with cleaners.

Torch Consumables and Wire Feed Setup

Aluminum wire is soft and low-melting, which makes it unusually sensitive to the feed system. A dirty liner or drive rolls set up wrong will snag the wire, cause bird-nesting, and produce an erratic deposit. Shops blame prep for that, so if the bead looks fine until the wire feeds badly, check the gun before the plate.

AWS’s Welding Digest identifies a spool gun or push-pull system as necessary for feeding soft aluminum wire without snarls in GMAW. Aluminum wire is far softer than steel, so a push-only setup can buckle and tangle in a long gun cable.

• Contact tips: size to the wire, replace when the bore wears or glazes. A worn tip spits or arcs off-center.
• Gun cable and liner: inspect often. Swarf, oxidized liner, and a kinked cable cause erratic feed.
• Drive rolls: use knurled or V-knurl rolls for soft aluminum wire, tensioned so the wire is not crushed.
• Shielding gas: argon for MIG and TIG unless a procedure calls for a helium mix. Purity and flow matter as much as anything in the torch.

Building the process from scratch? Our complete aluminum MIG welding guide covers gun setup and feed behavior in detail.

Protecting Finished Aluminum

Once the weld is done, the job is still exposed: keep cylinders, chains, clamps, and lifting gear away from finished aluminum. Steel chains leave rust and iron on the surface, and cutting aluminum near galvanized or stainless steel introduces zinc and iron contamination. Both cause porosity and cracking later, and both stay invisible until a weld blows holes.

Stage aluminum separately from steel, use dedicated slings and clamps, and do steel cutting and grinding in a different area or time than aluminum finishing.

Finally, the fume picture: NIOSH-listed hazards such as hexavalent chromium come mainly from stainless and chrome-bearing steels, not aluminum. That does not relax the need for local exhaust, because aluminum welding still produces oxide fume and ozone. More in the safety section.

Safety — Solvents, Fume, Oxide Dust and Ventilation

Cleaning aluminum is simple work, but the byproducts are not. Degreaser vapor, oxide dust, and aluminum fume each carry exposure limits and real fire risk, and striking an arc adds ozone and nitrogen oxides to the mix. A disciplined prep station is the cheapest safety upgrade any shop can make.

Solvents: Flammable, Ventilated, Disposed of Properly

Acetone, non-chlorinated brake cleaner, and dedicated aluminum degreasers are your most-used prep chemicals. All are flammable, and none belong next to hot work. Read the SDS before the first pour: it gives you the flash point, ventilation requirement, glove material, and disposal stream. Work where vapors can escape, cap the container when done, and keep ignition sources, such as a hot grinder or a cooling workpiece, out of the area. Our guide to picking an aluminum cleaner for welding compares the main types.

The most common mishap is not the solvent, it is the rag. Oily shop towels can self-heat and ignite without warning, so used wipes go straight into a closed, non-leaking container, and that container belongs in a metal can rated for oily rags. Not a plastic bag. Not an open bin.

The Fume and Gas Numbers

Striking an arc on aluminum generates aluminum oxide fume, and the arc’s ultraviolet output also produces ozone and nitrogen oxides. Under 29 CFR 1910.1000, Table Z-1, OSHA’s permissible exposure limit for welding fumes (total) is 5 mg/m³. For aluminum metal specifically, OSHA’s Occupational Chemical Database lists 15 mg/m³ for total dust and 5 mg/m³ for the respirable fraction, while NIOSH recommended exposure limits are 10 mg/m³ total and 5 mg/m³ respirable. The NIOSH Pocket Guide entry for aluminum (pyro powders and welding fumes, as Al) lists a REL of 5 mg/m³, with the skin and respiratory system as target organs and irritation and pulmonary fibrosis among the reported effects.

Aluminum fume behaves differently from steel fume. It is lighter, hangs in the air longer, and plates out on everything nearby, even vehicle paint outside. Extraction sized for a steel booth is often undersized for aluminum work.

Control Order: Ventilation First, PPE Last

The hierarchy matters more than any single piece of equipment. Engineering controls come first: general exhaust ventilation to clear the room, and local exhaust positioned close enough to the arc to capture the plume at the source. Then assess: is the hood actually moving enough air, does extraction run whenever the welder does, and does the capacity match the aluminum you are welding rather than the mild steel you welded yesterday? PPE is the last layer, never the first, and it only protects the person wearing it. OSHA’s welding fume and gas control resources spell out this sequence, and AWS’s Welding Digest makes the same case for aluminum, where ventilation is critical because the fumes carry ozone and nitrogen oxides alongside the oxide.

Oxide Dust: Vacuum, Never Sweep

Never dry-sweep aluminum dust or blow it off the worktable with compressed air. Fine aluminum dust is combustible, and blowing it away simply relocates the contamination onto the next joint you planned to weld. Use a HEPA-filtered vacuum, ideally one dedicated to aluminum so it never re-introduces steel dust, and never brush a hot surface. Any grinding, filing, or wire-brushing should happen with the dust captured at the tool.

PPE and Respiratory Protection

• Helmet: aluminum welding runs higher amperage for the same thickness, so move up in shade number rather than squinting through a lens that is too light.
• Clothing: flame-resistant clothing, sleeves down, and no synthetic cuffs near the arc.
• Face and eyes: safety glasses or a face shield when handling degreasers, since a splash plus a fume plume is a bad afternoon.
• Respirator: a supplement, not a substitute for ventilation, and only under a written program with fit testing and medical evaluation as required by 29 CFR 1910.134.

Containers, Cylinders and the Arc You Should Not Strike

Never put an arc on a container that held flammables, even one that looks empty and even after a rinse. Vapors trapped in the bottom flash back readily and can explode. Keep degreaser drums, thinners, and aerosol cans out of the welding bay, not just “away from the arc.” Verify that shielding gas cylinders are upright, chained, and capped, and check the regulator and hose for damage before connecting. Gas handling and cylinder safety details are covered in our guide to welding gas supply and cylinder safety.

Ozone, UV and the People Standing Next to You

AC GTAW and high-frequency starts add extra ultraviolet exposure, and AWS flags the high-frequency start flash as a hazard for personnel working nearby. That means UV-rated face shields for helpers, screens or curtains around the cell, and never a bare skin flash burn. For regulated work, the safety program is not optional: standards-driven programs such as those described in our aerospace welding guide assume documented cleaning, ventilation, and exposure procedures.

Troubleshooting — Porosity, Lack of Fusion, Cracks and Arc Instability

When an aluminum weld fails, it’s rarely a mystery. Most defects can be traced back to contamination, shielding, or settings. Of these, the two prep steps of degreasing and removing the oxide cover the majority of issues seen in the shop. Everything else comes down to gas, heat input, or your machine.

Quick Reference for Common Defects

Symptom	Likely cause	Fix
Round voids, worm-like holes, or bubbles at the root	Moisture: damp stock, wet filler, humid shop, or moisture in torch insulation	Dry stock and filler, re-degrease, check humidity and torch water traps
Gas pockets clustered at the start or stop	Oils and grease from shared rags, cutting oil, or fingerprints	Degrease with a fresh, single-use towel
Fine bright specks or a chalky skin inside the groove	Oxide remnant left in the bevel	Re-brush the bevel immediately before striking the arc
Porosity only when clamped	Zinc transfer from galvanized clamps or nearby steel grinding	Remove galvanized hardware and isolate all steel grinding in the area
Dull, black, poorly wetted bead with little penetration	Lack of fusion: oxide, narrow bevel, incorrect AC balance, or travel too fast	Re-clean, open the bevel, slow travel, and correct AC balance
Bright, jagged crack along the centreline	Hot cracking from filler and alloy pairing	Change filler to a suitable pairing for the base alloy and review joint design
Dull, wide crack appearing hours later	Contamination cracking from moisture or zinc	Re-degrease, re-brush, and eliminate all galvanized contact
Arc wandering, erratic, or a dirty colour	Contaminated tungsten, excess electrode-positive balance, or wrong tungsten type	Use a dedicated brush, fit a fresh tungsten, and verify AC balance
Crack at the end of the bead	Crater crack from rapid cool-down	Fill the crater by ramping down with a foot switch or taper function

Porosity: The Most Common Culprit

Porosity is one of the easiest defects to misdiagnose, as it looks like a heat issue when it’s almost always a cleanliness or shielding issue. Hydrogen is the driving factor here, as even the smallest amount of moisture can find its way into the molten aluminum pool and form round voids as it escapes.

Always work your way through sources in order. Start with the stock, filler rod, and any tools that may have moisture or coolant on them. From there, look at your torch for water traps, and finally your shielding gas. For TIG, aim for roughly 20 to 25 CFH of argon and check that flow at the torch, not just at the regulator. Low flow allows air to creep into the shield and create the same type of voids.

If you’ve fully re-prepped the joint, dried your filler, and the holes still persist, it’s time to look past prep. For setups that use wire feed, issues such as gas leaks, liner problems, or drive issues can mimic prep failures. This is especially true for Aluminum MIG setups, where those machine variables tend to rear their head faster.

Lack of Fusion

Lack of fusion occurs when the puddle simply won’t wet to the base metal. On aluminum, oxide is almost always the first suspect, particularly if it’s been left inside a tight bevel. From there, a bevel that is too narrow, excessive electrode-positive balance, travel that outruns the puddle, or a dirty tungsten can all prevent proper bonding.

If your bead is dull, black, and sluggish with no penetration, stop turning up the amps first. Go back to your surface prep. The aluminum cleaning guide on weldgallery is the perfect place to confirm your cleaning method before adjusting your heat input.

Cracking: Hot or Contamination-Based

Not all cracks are created equal. Hot cracking happens while the weld is still molten, most often along the centreline. This is almost always an alloy and filler pairing issue. Fillers with higher silicon content, such as 4043 or 405, can be crack-prone on 2xxx or 6xxx base alloys. In many of those cases, 5356 is the safer pairing for the application.

Cold, or contamination cracking, is a different story. These cracks tend to show up hours later, look wider and duller, and are caused by moisture or zinc. Unlike hot cracking, both of these are strictly prep problems and can be fully prevented with the correct cleaning and isolation steps.

Arc Instability and Crater Cracks

An arc that refuses to stay centered is a red flag for tungsten contamination. Contact with steel fixtures, dirty gloves, or using a steel brush will coat the tip and cause erratic wandering. Always use a dedicated aluminum brush and swap to a fresh pure or zirconiated tungsten at the first sign of discoloration or a rough tip. Excess electrode-positive balance can also destabilize the arc, and a helium mix will only add more heat to an already tricky puddle.

Lastly, never snap the arc off at the end of a bead. Crater cracks form as the puddle freezes with no filler to compensate for shrinkage. Use a foot switch or a ramp-down taper function to gently fade the arc out and fill the crater before breaking it.

How Prep Changes for TIG, MIG and Other Processes

Degreasing and oxide removal apply to every aluminum process. What changes is how much help you get: an arc can do part of the cleaning job, or none of it, and a resistance weld has no arc at all. Knowing who actually cleans the joint keeps you from prepping twice.

TIG (GTAW): The Arc Lends a Hand

GTAW on aluminum runs on alternating current with a high-frequency start and adjustable balance control, as the AWS Welding Digest describes for GTAW aluminum work. About half of every cycle is electrode positive, and that half does the work: the workpiece becomes the cathode, so the arc scours the surface and breaks the oxide back off. Miller’s guidance on AC balance control calls that cleaning action key to achieving a proper aluminum weld.

Balance is a dial, not a setting to max out. More electrode-positive time means more cleaning, but also faster tungsten wear and a less stable arc. Shielding stays at 100 percent argon with no mixes, with flow near 20 to 25 CFH.

Filler handling matters as much as prep: feed rod at the leading edge of the puddle from an oven or clean pipe, never with gloves used near a running machine.

MIG (GMAW): You Clean, the Machine Welds

Aluminum MIG runs on direct current with the electrode positive, normally labeled DCEP. Watch the naming: DCEN means direct current electrode negative, the opposite setting, and some shops just say “reverse polarity.” Check what your own machine calls it.

Electrode positive gives you the penetration and fusion aluminum MIG is known for, but no cathodic cleaning action. Whatever oxide is on the plate is still there when the wire touches it, so mechanical cleaning is not optional, and grease gets welded straight into the bead. As the Hobart Brothers aluminum guide puts it, porosity in aluminum requires that all moisture and hydrocarbons be eliminated before welding.

Feed soft aluminum wire with a spool gun or push-pull gun so it stays straight, and use the push technique with the torch trailing the puddle rather than drag. Dragging stubs the wire and disturbs a fragile puddle. A kinked liner or worn drive roll produces bird-nesting and an erratic bead often blamed on prep. Our aluminum MIG welding guide covers gun setup, cable condition and polarity.

Process	Where cleaning comes from	Filler handling	Defect signature
TIG / GTAW	AC electrode-positive half cycle plus the brush	Oven or clean pipe, rod at the leading edge of the puddle	Crater cracks, dark bead, damp-rod porosity
MIG / GMAW	Mechanical prep only	Spool gun or push-pull, push technique	Bird-nesting, erratic deposit, cold lap, porosity
Brazing with flux	Flux dissolves what is left, at lower temperature	Pre-placed filler or drawn through the capillary gap	Pitting, incomplete wetting, oxide islands
Resistance welding	Wiping action of electrode tips, cleaned often	None	No nugget, expulsion, inconsistent resistance
Laser welding	None, the beam reacts to the finish you leave	Optional wire feed, kept clean	Porosity, unstable keyhole, inconsistent depth

Brazing, Resistance, Laser and Cutting Prep

Brazing aluminum happens below the base metal’s melting point, so flux does more of the work than on a weld. Skip the mechanical cleaning and flux has far more oxide to chew through, which shows up as pitting and poor wetting. Resistance welding is where clean metal matters most, because aluminum oxide is an insulator and every layer left on the surface drives contact resistance up, exactly what the process tries to control. Clean the electrode tips on a schedule too.

Laser welding is sensitive to finish in a way most welders do not expect: a bright, freshly machined surface reflects energy differently from a matte, lightly abraded one, so identical settings behave differently depending on what you cleaned to. Change your prep method and expect to re-tune; our alpha laser welding overview covers the equipment side. Plasma cutting and arc gouging are prep for the cut, not the weld: gouge back to sound metal, clean the dross and heat-affected edge, and remove the hard dark layer before welding over it.

Hot cracking is the classic misdiagnosis. A crack down the center of a clean, bright, prepped bead usually points to an alloy and filler pairing issue, not a degreaser issue, so check the filler spec before you re-clean anything. For structural and aerospace work, AWS D1.2/D1.2M and AWS D17.1 assume qualified cleaning procedures, not guesswork; you can find them on the AWS standards listings. If your work is certified, the prep method becomes part of the procedure, so write it down and train the shop to match. The aerospace TIG welding guide shows what that qualification demands.

A Repeatable Prep Checklist and the Mistakes That Cost You Most

Everything so far collapses into one thing: a sequence you can run without thinking. Most aluminum weld failures are not exotic. They are a skipped wipe, a brush that touched steel last week, a tack made forty minutes before the real weld. A checklist turns prep from memory into habit, and habit is what survives a hot Friday when the fit-up fights you.

Run this on every aluminum job, start to finish. Once you have it down it takes under five minutes.

1. Identify the alloy and thickness. Mark the joint so others read the note. It decides filler metal, bevel, and heat input.
2. Confirm the filler matches the job. Check rod or spool against alloy and service conditions, not what is on the bench.
3. Remove oils. Degrease with a suitable solvent and a clean, single-use towel. Wipe one direction, change towels often.
4. Look for white or chalky oxide. A dull, chalky, gritty surface needs a dedicated stainless brush, across the grain, light pressure.
5. Verify edge prep and fit-up. Bevel angle, land, root gap, clamping. Consistent gaps beat clever fixturing.
6. Check the filler rod or wire. Clean, dry, oil free. Rod from a dusty box stays out of the joint.
7. Check shielding gas and flow. Correct cylinder, correct flow rate, no leaks in line or torch. Argon, no exceptions.
8. Check the tools. Brushes, files, and grinding media must be aluminum-dedicated, free of steel carryover. Check tip, liner, drive roll.
9. Tack with the same prep and filler you intend for the final weld, on the same side as the final pass.
10. Wipe the joint immediately before welding. Final clean, single-use towel, then strike up within minutes. Oxide returns as soon as metal meets air.

The Mistakes That Cost the Most

These are the errors that keep showing up in failed aluminum joints. Most are recoverable the first time you find them. Repeat them often enough and the joint goes to the scrap pile.

Mistake	What It Causes	Fix
Steel brush	Iron carryover, tungsten contamination, erratic arc	Stainless brush, aluminum only, stored separately
Same rag for aluminum and steel	Lanolin, detergent residue, iron transfer	White paper towels or fresh single-use towels
Degreasing after grinding	Abrasive and oil residue ground into the surface	Degrease, clean, then mechanically clean
Skipping bevel cleanup	Chalky oxide survives inside the groove	Brush bevel faces and root land before tacking
Welding over a wet or greasy tack	Contaminant trapped where no wipe can reach	Degrease before tacking, keep tacks clean, re-prep if handled
Oily gloves on the plate	Fingerprints and skin oils at the joint	Clean dedicated gloves, handle prepared areas only
Filler rod in a dusty box	Porosity in a perfectly prepped joint	Rod oven or clean closed pipe, dry and covered
Grinding wheel used on steel	Embedded iron, then a contaminated bead	Dedicated aluminum grinding tools only
Chlorinated or silicone-laden cleaners	Corrosion and residue that survive the arc	Acetone, non-chlorinated brake cleaner, aluminum degreaser
Waiting too long after prep	Fresh oxide and moisture re-form	Clean right before welding, not at shift start

One more worth naming because it is easy to miss: silicone. Many penetrating sprays, sealants, and polishes contain it. Once silicone reaches aluminum it is nearly impossible to remove, and it will drive porosity and cracking for the life of the joint. If a joint keeps failing for no reason, find out what has been sprayed nearby.

When to Rework and When to Scrap

Most failed aluminum welds are recoverable: grind out the bad bead completely, remove all ground material, degrease, mechanically re-clean the surface and adjacent parent metal, and run the joint again with fresh filler and verified gas. Porosity often re-welds in place, but a hot crack must be cut out entirely, because a crack left in the joint keeps growing under load. Scrap the joint when the base metal is damaged beyond repair, when a crack ran deep into the plate or extrusion, when the weld sits in a load path you cannot qualify a rework, or when contamination cannot be fully removed. If you cannot identify why the joint failed, do not weld it again and hope. Find out first.

The stakes are not cosmetic. ESAB/AlcoTec prep guidance is blunt: in structural or high-load applications, porosity, lack of fusion, arc instability, and contamination cracking can lead to catastrophic failure. A bead that chews apart under a chip hammer is not a cosmetic problem. It is a weld that never fused.

Prep is not the only variable. If you run MIG on aluminum, the gun, liner, and drive roll setup deserve their own review before you blame the joint. The Aluminum MIG welding guide covers that setup in detail.

One last habit. If a joint is giving you trouble and you are not sure why, change one thing, and change the prep first. Clean it, degrease it, brush it immediately before welding, and watch what happens. Aluminum does not hide its problems, and neither should your process.

Frequently Asked Questions

Do I really have to clean aluminum before welding?

For aluminum, cleaning is not optional polish — it is the part of the process that makes fusion possible. Because the oxide layer melts at roughly 3,600°F while the base metal melts near 1,220°F, leaving it in place produces porosity, lack of fusion, and arc instability. New, clean, dry material may get away with degreasing alone, but any stained, chalky, or weathered surface must be mechanically cleaned.

What is the best way to clean aluminum before welding — solvent, wire brush, or both?

Both, in that order. Degrease first with acetone, a non-chlorinated brake cleaner, or a mild alkaline cleaner, then remove the oxide mechanically with a dedicated stainless steel wire brush, file, or hand planer. ESAB’s guidance is that if you can only do one step, degrease, because oils pose the greatest threat — but combining both steps always gives the highest level of assurance.

Can I use a regular steel wire brush on aluminum?

No. Use a dedicated stainless steel brush and keep it in a separate container used only for aluminum. A steel brush leaves iron particles on the surface, which can contaminate the weld, cause porosity, and damage your tungsten. Label the brush, hang it apart from steel brushes, and replace it when it becomes loaded.

Is acetone the best degreaser for aluminum?

Acetone is a widely used choice because it dissolves oil, cuts through moisture, and evaporates cleanly without leaving a film. Non-chlorinated brake cleaners and mild alkaline cleaners also work, and a household dish soap in hot water is a reasonable option for hobbyists if you rinse thoroughly with hot water. Avoid chlorinated solvents, silicones, kerosene, and heavy oils.

How soon before welding must I clean the aluminum?

As close to welding as practical. Aluminum begins re-forming its oxide the moment it is exposed to air, so clean the joint immediately before you tack or strike an arc, and clean each side of the bevel rather than only the top face. Do not touch the cleaned surface with bare or greasy hands — use clean gloves.

Why do my aluminum welds have holes in them?

Porosity in aluminum is almost always moisture or contamination, and occasionally shielding gas. Work through it in order: is the plate oily or did someone handle it with dirty gloves, is there white chalky oxide left in the groove, is the filler rod or wire damp or dusty, is there zinc contamination from a galvanized clamp or nearby grinding, and is your argon flow high enough to keep air out of the pool?

What causes lack of fusion in aluminum welds?

Lack of fusion usually means the oxide was still present when the arc arrived, or the heat input was too low to melt the base metal properly. Check that the joint was brushed immediately before welding, confirm the bevel is clean, and verify your AC balance or DC polarity settings for the machine. On MIG, make sure you are pushing rather than dragging the torch and that the wire is feeding smoothly.

Does welding aluminum require preheating?

For most aluminum alloys and most thicknesses, no — preheat is usually unnecessary and can actually increase distortion. Where a qualified procedure specifies a minimum interpass temperature, follow the procedure. Manage heat with a good fit-up, a stable travel speed, stringer beads, and occasionally a clean sacrificial backing bar rather than with a torch preheat.

Are aluminum welding fumes dangerous?

They should be treated seriously. OSHA lists a permissible exposure limit of 15 mg/m³ for total aluminum dust and 5 mg/m³ for the respirable fraction, NIOSH lists a 5 mg/m³ REL for aluminum welding fumes, and total welding fume is limited to 5 mg/m³ under 29 CFR 1910.1000. Welding aluminum also produces ozone and nitrogen oxides, so local exhaust and good general ventilation are the first control, with respirators used only as a supplement under a written program.

Do I need to clean the aluminum filler rod and wire too?

Yes. Store aluminum rod and wire clean, dry, and protected — in a rod oven or a clean closed pipe — and handle them with clean gloves. A contaminated or damp filler rod can put porosity in the middle of an otherwise perfectly prepped joint, which sends troubleshooting in the wrong direction. Also keep consumables and tools dedicated to aluminum.